镜像自地址
https://github.com/binary-husky/gpt_academic.git
已同步 2025-12-07 15:06:48 +00:00
refine
这个提交包含在:
@@ -34,17 +34,6 @@ def extract_text(file_path):
|
||||
# 使用 SimpleDirectoryReader 处理它支持的文件格式
|
||||
if ext in ['.docx', '.doc']:
|
||||
return read_docx_doc(file_path)
|
||||
else:
|
||||
# 尝试读取文件为纯文本
|
||||
try:
|
||||
with open(file_path, 'r', encoding='utf-8') as f:
|
||||
return f.read()
|
||||
except UnicodeDecodeError:
|
||||
# 如果遇到解码错误,说明文件不是纯文本
|
||||
print(f"Error: {file_path} is not a valid text file.")
|
||||
except Exception as e:
|
||||
# 捕获其他可能的异常
|
||||
print(f"Error reading file {file_path}: {e}")
|
||||
try:
|
||||
reader = SimpleDirectoryReader(input_files=[file_path])
|
||||
documents = reader.load_data()
|
||||
|
||||
@@ -15,6 +15,8 @@ def 文档总结(file_manifest, project_folder, llm_kwargs, plugin_kwargs, chatb
|
||||
file_content = extract_text(fp)
|
||||
# private_upload里面的文件名在解压zip后容易出现乱码(rar和7z格式正常),故可以只分析文章内容,不输入文件名
|
||||
if file_content==None:
|
||||
chatbot.append(
|
||||
[f"上传文件: {os.path.basename(fp)}", f"此文件解析失败,无法提取文本内容。失败原因可能为:1.文档格式过于复杂;2. 不支持的文件格式,支持的文件格式后缀有:" + ", ".join(supports_format)+ "等其他文本格式类型文件。"])
|
||||
continue
|
||||
from crazy_functions.pdf_fns.breakdown_txt import breakdown_text_to_satisfy_token_limit
|
||||
from request_llms.bridge_all import model_info
|
||||
@@ -55,13 +57,11 @@ def 文档总结(file_manifest, project_folder, llm_kwargs, plugin_kwargs, chatb
|
||||
|
||||
res = write_history_to_file(history)
|
||||
promote_file_to_downloadzone(res, chatbot=chatbot)
|
||||
chatbot.append(("完成了吗?", res))
|
||||
chatbot.append((f"路径{fp}文件解读完成了吗?", "解读完成,存储路径为"+res))
|
||||
yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
|
||||
|
||||
res = write_history_to_file(history)
|
||||
promote_file_to_downloadzone(res, chatbot=chatbot)
|
||||
chatbot.append(("所有文件都总结完成了吗?", res))
|
||||
yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
|
||||
|
||||
|
||||
@CatchException
|
||||
|
||||
在新工单中引用
屏蔽一个用户